STRUCTURE ADAPTABLE DIGITAL NEURAL NETWORK TECHNIQUES IN VISUAL SERVOING OF ROBOTIC MANIPULATORS

Partap Singh, Harvir Singh

Abstract: Robotics is one of the most challenging applications of soft computing techniques. It is characterized by direct interaction with a real world, sensory feedback and a complex control system. This paper reviews the application of soft computing approaches, particularly neural networks, in the domain of visual servoing of robotic manipulators. Various robotic tasks within the scope of visual servoing are identified, and the issues involving the application of soft computing approaches in solving these problems are discussed. The paper provides some practical suggestions in applying neural networks for these tasks.

Keywords: NEURAL NETWORK

DOI: https://doi.org/10.15623/ijret.2014.0326005